Group B, Poster #094, Earthquake Geology

Pre-rupture fault mapping from geomorphic features to quantify fault location uncertainty for fault displacement hazard studies

Malinda G. Zuckerman, Chelsea P. Scott, Ramon Arrowsmith, Rachel N. Adam, Rich D. Koehler, Christopher M. Madugo, & Albert R. Kottke
Poster Image: 

Poster Presentation

2022 SCEC Annual Meeting, Poster #094, SCEC Contribution #12270 VIEW PDF
Fault maps indicate hazards associated with fault displacement that can impact structures and lifelines that cross the faults. Earthquakes leave geomorphologic traces in the landscape where, within aleatoric variability, the traces may rupture again. It is standard practice to map and classify these features to identify the location of future ruptures. To test the indicative power of the features to predict the location of surface deformation, we created pre-rupture fault maps and tested them against co-seismic ruptures.

Our team developed and taught a virtual course in Spring 2022. We taught students how to make fault maps based on topographic and imagery datasets acquired be...
fore several earthquakes. Students ranged from senior undergraduates to post doctoral scholars from five universities in three countries. We implemented a systematic approach to fault mapping involving mapping surficial geology, morphology, and tectonic geomorphic indicators before delineating fault traces. The class focused on two major activities:

1. All students mapped the same areas independently, addressing mapping repeatability. If faults can be mapped similarly by multiple people, it suggests the signals in the landscape are recognizable and significant. We examined how experience level affects mapping. Inexperienced students introduce epistemic uncertainty due to misunderstanding the geomorphology. They do not map with the prior knowledge of a more experienced mapper. The width of the uncertainty surrounding fault traces is indicated by the features multiple mappers cite while mapping the same fault. Over half of the students anticipated the trend of the coseismic ruptures in four out of five mapping areas. Students anticipated rupture trace segments within 150 m.

2. Projects addressed the impacts of mapping faults using different types of base datasets. Data types varied in quality and included datasets such as 30m/pix SRTM data, historical aerial imagery, 1-2 m/pix lidar, and 10cm/pix sUAS data processed with SfM. We sought to discover what makes a dataset helpful to mapping faults and what kind of dataset is most appropriate to reveal geomorphic indicators. Students were most successful when mapping with higher resolution topographic datasets. Aerial imagery was limiting in areas with high vegetation and anthropogenic alteration, but helpful as a supplement to topographic data.